Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Biochemical Characterization of CTX-M-15 from Enterobacter cloacae and Designing a Novel Non-β-Lactam-β-Lactamase Inhibitor

  • Mohammad Faheem ,

    Contributed equally to this work with: Mohammad Faheem, Md Tabish Rehman

    Affiliation Medical Microbiology and Molecular Biology Laboratory, Interdisciplinary Biotechnology Unit, Aligarh Muslim University, Aligarh, UP, India

  • Md Tabish Rehman ,

    Contributed equally to this work with: Mohammad Faheem, Md Tabish Rehman

    Affiliation Medical Microbiology and Molecular Biology Laboratory, Interdisciplinary Biotechnology Unit, Aligarh Muslim University, Aligarh, UP, India

  • Mohd Danishuddin,

    Affiliation Medical Microbiology and Molecular Biology Laboratory, Interdisciplinary Biotechnology Unit, Aligarh Muslim University, Aligarh, UP, India

  • Asad U. Khan

    asad.k@rediffmail.com

    Affiliation Medical Microbiology and Molecular Biology Laboratory, Interdisciplinary Biotechnology Unit, Aligarh Muslim University, Aligarh, UP, India

Correction

30 Dec 2013: Faheem M, Rehman MT, Danishuddin M, Khan AU (2013) Correction: Biochemical Characterization of CTX-M-15 from Enterobacter cloacae and Designing a Novel Non-β-Lactam-β-Lactamase Inhibitor. PLOS ONE 8(12): 10.1371/annotation/049bf1aa-d866-471f-95c1-5939d4461f8c. https://doi.org/10.1371/annotation/049bf1aa-d866-471f-95c1-5939d4461f8c View correction

Abstract

The worldwide dissemination of CTX-M type β-lactamases is a threat to human health. Previously, we have reported the spread of blaCTX-M-15 gene in different clinical strains of Enterobacteriaceae from the hospital settings of Aligarh in north India. In view of the varying resistance pattern against cephalosporins and other β-lactam antibiotics, we intended to understand the correlation between MICs and catalytic activity of CTX-M-15. In this study, steady-state kinetic parameters and MICs were determined on E. coli DH5α transformed with blaCTX-M-15 gene that was cloned from Enterobacter cloacae (EC-15) strain of clinical background. The effect of conventional β-lactamase inhibitors (clavulanic acid, sulbactam and tazobactam) on CTX-M-15 was also studied. We have found that tazobactam is the best among these inhibitors against CTX-M-15. The inhibition characteristic of tazobactam is defined by its very low IC50 value (6 nM), high affinity (Ki = 0.017 µM) and better acylation efficiency (k+2/K′ = 0.44 µM−1s−1). It forms an acyl-enzyme covalent complex, which is quite stable (k+3 = 0.0057 s−1). Since increasing resistance has been reported against conventional β-lactam antibiotic-inhibitor combinations, we aspire to design a non-β-lactam core containing β-lactamase inhibitor. For this, we screened ZINC database and performed molecular docking to identify a potential non-β-lactam based inhibitor (ZINC03787097). The MICs of cephalosporin antibiotics in combination with this inhibitor gave promising results. Steady-state kinetics and molecular docking studies showed that ZINC03787097 is a reversible inhibitor which binds non-covalently to the active site of the enzyme through hydrogen bonds and hydrophobic interactions. Though, it’s IC50 (180 nM) is much higher than tazobactam, it has good affinity for CTX-M-15 (Ki = 0.388 µM). This study concludes that ZINC03787097 compound can be used as seed molecule to design more efficient non-β-lactam containing β-lactamase inhibitor that could evade pre-existing bacterial resistance mechanisms.

Introduction

Antibiotic resistance in Gram-negative bacteria is a major health concern. It is principally observed due to the emergence of β-lactamase producers, which leads to the resistance against β-lactam antibiotic [1]. The β-lactamase enzymes are classified according to the scheme of Ambler into four classes, designated classes A to D, on the basis of their amino acid sequences, with classes A and C being the most frequently occurring among bacteria. Extended Spectrum β-lactamases (ESBLs) belong to molecular Amber class A or functional class 2be β-lactamases capable of conferring bacterial resistance to the penicillins, narrow-spectrum, expanded-spectrum and broad spectrum cephalosporins, and aztreonam, but not to cephamycins or carbapenems. Moreover, their activity is inhibited by β-lactamase inhibitors such as clavulanic acid, sulbactam and tazobactam [2], [3].

Plasmid-encoded ESBLs of the CTX-M type are increasingly being reported worldwide in Gram-negative bacteria and now account for most of the ESBL types found in the Enterobacteriaceae [4], [5]. They form a rapidly growing family that comprises more than 100 variants (http://www.lahey.org/studies) and are divided into five groups according to amino acid sequence identity, with different groups being prevalent in different countries [6], [7]. CTX-M ESBLs (derived its name from being highly active on CefoTaXime and isolated in Munich) are characterized by displaying greater hydrolytic activity against cefotaxime than against ceftazidime [8]. However, some clinical isolates have a significant degree of resistance to ceftazidime as well [9]. The first clinical strain producing CTX-M enzymes was found in Japan in 1993 with the characterization of the Toho-1 enzyme from an E. coli strain [10]. CTX-M-3 and its variants, CTX-M-15 were discovered in Poland, India, United Kingdom, Bulgaria, Romania and Turkey [9], [11], [12]. In India, CTX-M-15 is the most widespread ESBL and has been reported from six unrelated members of the family Enterobacteriaceae (four E. coli strains, one K. pneumonia strain, and one E. aerogenes strain) [9]. The widespread dissemination of CTX-M-15 has a significant impact on the treatment of hospital- and community-acquired infections caused by E. coli and other enteric bacilli [13][15].

Several β-lactamase inhibitors that are commonly used in combination with β-lactam antibiotics are clavulanic acid, tazobactam and sulbactam. Among class A enzymes, tazobactam is the most potent inhibitor followed by clavulanic acid and sulbactam [16]. The core structure of these inhibitors contains a β-lactam ring (Figure 1). Emergence of bacterial resistance against such inhibitors has been reported owing to the ability of bacteria to hydrolyse the β-lactam core of these inhibitors [17][19]. Porin channel mutation and overexpression of β-lactamases in the presence of β-lactam based inhibitor are other mechanisms that confer increasing resistance against such inhibitors [20]. Thus, there is an urgent need for the screening of novel inhibitors that do not contain a β-lactam core structure. Such inhibitors would not be hydrolyzed by wild type or mutant β-lactamases and would not be recognized by the ESBL producers [21]. Moreover, a novel non-β-lactam based inhibitor would not be affected by porin channel mutations, which prevent β-lactams from accessing their cellular targets. Furthermore, non-β-lactam based inhibitors would minimize the ability of bacteria to recruit existing resistance mechanisms, and bacteria would take a long time to develop novel mechanisms of resistance [22].

thumbnail
Figure 1. Chemical structure of β-lactamase inhibitors.

Different inhibitors used in the study are (a) clavulanic acid, (b) tazobactam, (c) sulbactam, and (d) ZINC03787097.

https://doi.org/10.1371/journal.pone.0056926.g001

Previously, we have identified CTX-M-15 from E. coli, E. cloacae, K. pneumoniae and A. Baumanii from Aligarh hospital of north India and submitted their DNA sequences in Genbank [13], [23]. In the present study, blaCTX-M-15 from an Enterobacter cloacae clinical strain, EC-15 (Genbank accession no.: JN860195.1) was cloned and the enzyme was purified to homogeneity and an attempt has been made to understand the correlation between MICs and catalytic activity. This study also aimed to identify novel non-β-lactam core containing inhibitor and explore its mechanism of action.

Materials and Methods

Antibiotics and Other Chemicals

Ampicillin, Piperacillin, Cefazolin, Cefuroxime, Cefotaxime, Ceftriaxone, Ceftazidime, Cefepime and Aztreonam were purchased from Sigma chemical co. (St. Louis, MO), and Nitrocefin was purchased from Calbiochem (USA). Clavulanic acid, Sulabctam and Tazobactam were from Sigma-Aldrich (St. Louis, MO), while ZINC03787097 was purchased from Santa Cruz, India. IPTG (isopropyl-β-D-thiogalactopyranoside) was purchased from Roche (Basel, Switzerland). Other reagents and chemicals were of analytical grade. The structures of various inhibitors used in the present study were presented in figure 1.

Bacterial Strains

E. coli DH5α and E. coli BL21 (DE3) were used for cloning and protein expression experiments, respectively. MICs were determined on E. coli DH5α transformed with cloned CTX-M-15 from Enterobacter cloacae clinical strain EC-15.

Cloning and Sequencing

The plasmid DNA harbouring blaCTX-M-15 gene from clinical E. cloacae EC-15 strain (Genbank accession no.: JN860195.1), characterized in our lab, was extracted using Qiagen plasmid extraction kit, according to manufacturer’s instructions. The blaCTX-M-15 gene was amplified by PCR with the primers CTX-M-15-F (5′ ATATCATATGGTTAAAAAATCACTG 3′) containing Nde I site and CTX-M-15-R (5′ ATATAAGCTTTTACAAACCGTCGGTGAC 3′) containing Hind III site. The PCR conditions used were 95°C (30 s), 54°C (25 s), 72°C (40 s) and the reaction was carried out for 35 cycles. The PCR product does not contain the promoter region of the gene. The PCR product and pQE-2 (high copy cloning vector), were double digested with Nde I and Hind III, ligated and used to transform competent E. coli DH5α by heat shock method. Transformants harbouring blaCTX-M-15 gene were selected on LB agar plates containing ampicillin (100 µg/ml). The clones were confirmed by sequencing on both strands by standard procedures.

CTX-M-15 β-lactamase Expression and Purification

To express and purify CTX-M-15 β-lactamase, the pQE-2 vector harbouring blaCTX-M-15 gene was transformed into competent E. coli BL21 (DE3) cells. A 5 ml overnight culture of these transformed cells in Luria-Bertani (LB) medium containing 100 µg/ml ampicillin was used to inoculate 1 litre of LB medium containing 100 µg/ml ampicillin. Bacteria were cultured at 37°C with shaking, until an optical density at 600 nm of 0.6 was reached. The culture was cooled and then transferred to 37°C, induced by 0.5 mM IPTG for three hours. The bacteria were collected by centrifugation and resuspended in 20 ml lysis buffer containing 50 mM Tris, pH 8.0, 300 mM NaCl and 0.1% β-mercaptoethanol per litre culture. The bacteria were ruptured by sonication, and the cell debris was removed by centrifugation at 12,000 rpm for 30 min. The cleared supernatant was loaded onto a Ni-NTA column, which was pre-equilibrated by lysis buffer, and washed with lysis buffer supplemented with 20 mM imidazole. Protein was eluted with PBS buffer containing 250 mM imidazole. Pure protein was obtained after dialysis in PBS (50 mM sodium phosphate, pH 8.0, containing 300 mM NaCl). Purity of the purified protein was estimated to be more than 95% as determined by a single band of 31 kDa on SDS-PAGE (Figure S1). The final protein concentrations were determined by using the molar extinction coefficient of 25, 440 M−1 cm−1 at 280 nm.

Antibiotic Susceptibility Testing Based on MIC Analysis

The MICs for various antibiotics alone or in combination with various inhibitors (clavulanic acid, sulbactam, tazobactam and ZINC03787097) were determined by micro-dilution method [24] and the results were interpreted according to Clinical Laboratory Standards Institute (CLSI) guidelines [25]. Briefly, E. coli DH5α cells transformed with CTX-M-15 was treated with increasing concentrations of the antibiotics ranging from 0.5 to 2000 µg/ml in a series of two fold dilutions. The inhibitors clavulanic acid, sulbactam and tazobactam were used at a fixed concentration of 4 µg/ml, while ZINC03787097 was used at 20 µg/ml. The MIC was determined as the lowest concentration that totally inhibits visible bacterial growth.

Determination of IC50

Various concentrations of clavulanic acid, tazobactam, sulbactam and ZINC03787097 were pre-incubated with the purified CTX-M-15 for 5 mins at 30°C before the rate of nitrocefin (100 µM) hydrolysis was measured [26], [27]. The 50% inhibitory concentration (IC50) was determined as the concentration of the inhibitor that inhibited hydrolytic activity of the enzyme by 50%.

Steady-state Kinetics Experiments

The steady-state kinetics parameters of CTX-M-15 were determined on the following antibiotics: Ampicillin (Δε235 = −900 M−1 cm−1), Piperacillin (Δε232 = −1640 M−1 cm−1), Nitrocefin (Δε486 = +15,000 M−1 cm−1), Cefazolin (Δε320 = +1067 M−1 cm−1), Cefuroxime (Δε262 = −8,540 M−1 cm−1), Cefotaxime (Δε264 = −7,250 M−1 cm−1), Ceftazidime (Δε265 = −10,300 M−1 cm−1), Cefepime (Δε267 = −9,120 M−1 cm−1) and Aztreonam (Δε318 = −650 M−1 cm−1). The inhibitors used were Clavulanic acid, Sulabctam, Tazobactam and ZINC03787097.

Hydrolysis of β-lactam antibiotics was detected by monitoring the variation in the absorbance due to cleavage of β-lactam ring in 50 mM phosphate buffer, pH 7.0 [28]. All the measurements were taken in triplicate on Shimadzu UV-VIS spectrophotometer (UV-1800). The reaction was performed in a total volume of 500 µl at 30°C. For dilution of the enzyme and to prevent denaturation, BSA was added to a final concentration of 20 µg/ml.

The kinetic parameters (kcat and Km) for the hydrolysis of good substrates (nitrocefin, ampicillin, piperacillin, cefazolin, cefuroxime, cefotaxime, and ceftriaxone) were determined by measuring the initial rate of antibiotic hydrolysis and by using Michaelis-Menten equations 1 and 2 [29].(1)(2)where, v and Vmax are the initial and maximum velocity of hydrolysis, respectively, [S] is the concentration of the substrate used, [E] is the enzyme concentration in the reaction, and Km is the Michaelis-Menten constant.

For poor substrates (ceftazidime, cefepime and aztreonam) and inhibitors (clavulanic acid, sulbactam, tazobactam, and ZINC03787097), the kcat values were determined from the initial rates calculated at saturating substrate concentrations (i.e. under zeroth order kinetics), and the Km values were determined as competitive inhibition constant (Ki) in a competition experiment between tested antibiotic/inhibitor and 100 µM nitrocefin used as reporter substrate, and the result was analysed according to equation 3 [28].(3)where, vo and vi represent the initial rate of nitrocefin hydrolysis in the absence and presence of the poor substrate, respectively; [C] is the concentration of poor substrate/inhibitor; and Km and [S] are the Michaelis-Menten constant and concentration of nitrocefin, respectively.

Evaluation of enzyme inhibition kinetic results.

The interaction of β-lactamases with inhibitors is represented by a simple three-step acylation-deacylation mechanism [30][32] as shown:(4)where, E is enzyme, C is inhibitor, EC is non-covalent Henri-Michaelis complex, EC* is acyl-enzyme intermediate and P is inactive degradation product of the substrate. The parameters k+2, k+3 and K′ are the first-order acylation and deacylation constants, and the dissociation constant of the Henri-Michaelis complex, respectively. The steady-state kinetic parameters for the above equation are given by the following equations:

(5)(6)(7)(8)

The values of the first-order rate constant (ki) characterizing the rate of EC* accumulation was calculated by measuring the hydrolysis of nitrocefin (reporter substrate) at different inhibitor concentrations according to the following equation [28]:(9)

where, vo, vt and vss are the rates of utilization of the reporter substrate at times 0 and t, and after the steady-state has been established, respectively.

The individual parameters, k+2, k+3 and K′, were derived from the dependence of ki upon inhibitor concentration, [C], on the basis of the following equation:(10)

where, Km and [S] are the Michaelis-Menten constant and concentration of the reporter substrate, nitrocefin.

When the concentration of inhibitor [C] is well below K′, ki varies linearly the above equations is simplified as:(11)

The k+2/K′ value is then obtained from the slope of the line, whereas k+3 is given by the intercept at [C] = 0. Conversely, if [C] is much larger than K′ (Km+[S])/Km, then ki = k+2+k+3, and is independent of [C] [28].

Mechanism of Action of ZINC03787097

The mode of action of ZINC03787097 was determined by monitoring the hydrolysis of nitrocefin (10–100 µM) by CTX-M-15 (0.50 nM) in the presence of different concentration of ZINC03787097 (0.25–0.75 µM) in 50 mM phosphate buffer, pH 7.0 at 30°C [28]. All the measurements were taken in triplicate on Shimadzu UV-VIS spectrophotometer (UV-1800). The result was analysed from Lineweaver-Burg plot.

Stability of ZINC03787097 in the Presence of CTX-M-15

The UV absorption spectrum (320-210 nm) of ZINC03787097 (100 µM) alone or after 3600 seconds incubation with CTX-M-15 (10 nM) was measured in 50 mM phosphate buffer, pH 7.0 at 30°C on Shimadzu UV-VIS spectrophotometer (UV-1800). The stability of ZINC03787097 in the presence of CTX-M-15 was monitored by measuring the absorbance for 3600 seconds at 283 nm and 222 nm.

In silico Modelling and Molecular Docking

Protein sequence of CTX-M-15 was retrieved from NCBI (ID: ABM88811). Three dimensional structure of this protein was built by using the template (pdb:1IYS), having 86% identity. The selected model was refined by performing energy minimization through a CHARMm force field [33] with Dependent Dielectric implicit solvent model and conjugates gradient method. This process was carried out until the average absolute derivative of co-ordinates with respect to energy fell below 0.1 kcal Å−1. Molecular docking of known β-lactamase inhibitors (clavulanic acid, sulbactam and tazobactam) along with screened inhibitor (ZINC03787097) from ZINC database [34] was performed by using GOLD (Genetic Optimization for Ligand Docking) 5.0 program [35]. GOLD fitness score was used as a parameter to define the efficiency of selected inhibitors.

Results

Antibiotic Susceptibility Testing Based on MIC Analysis

The MICs of the β-lactam antibiotics alone or in combination with different inhibitors (clavulanic acid, sulbactam, tazobactam and ZINC03787097) were determined on E. coli DH5α harbouring blaCTX-M-15 gene from Enterobacter cloacae isolate (EC-15) of clinical background and the results are presented in table 1. Very high MICs (500–1000 µg/ml) were obtained for ampicillin, piperacillin, cefazolin, cefuroxime, cefotaxime and ceftriaxone, indicating that the studied strain is highly resistant to these antibiotics. The MICs of ceftazidime, cefoxitin, aztreonam and cefepime (4, 16, 16 and 125 µg/ml, respectively) were moderate, but still is in the resistant range.

thumbnail
Table 1. MICs of β-lactams alone or in combination with inhibitors for E. coli DH5α transformed with recombinant blaCTX-M-15 from Enterobacter cloacae.

https://doi.org/10.1371/journal.pone.0056926.t001

The efficacy of antibiotic-inhibitor combination was also studied on DH5α transformed with blaCTX-M-15 gene (Table 1). It was found that none of the inhibitors considered in the study were able to reduce the MIC of ampicillin. The MIC of third generation cephalosporin cefotaxime was lowered 16, 2, 8 and 2 fold when the antibiotic was used in combination with clavulanic acid, sulbactam, tazobactam and ZINC03787097, respectively. The MIC of ceftazidime in combination with various inhibitors was reduced 2-folds to 2 µg/ml, and bought down to the susceptible range. Similarly, the MIC of cefepime was reduced to 2–16 folds in the presence of various inhibitors. The greatest effect of inhibitors on MIC was observed when they are given in combination with cefoxitin. The MIC was reduced from 16 µg/ml to 4 µg/ml for cefoxitin-sulbactam combination, and to 2 µg/ml in the case of cefoxitin with clavulanic acid, tazobactam and ZINC03787097.

IC50 Determination

The IC50 values are appropriate for accessing the potency of an inhibitor or comparing the potential of different inhibitors under properly controlled experiments. The IC50 values were determined by exposing CTX-M-15 to different inhibitors, namely clavulanic acid, sulbactam, tazobactam and ZINC03787097 for 5 mins and measuring the percent residual enzyme activity on nitrocefin (100 µM). The result is presented in figure 2 and table 2.

thumbnail
Figure 2. Determination of IC50 values for different inhibitors.

The residual activity of CTX-M-15 after 5 minutes pre-incubation with varying concentrations of different inhibitors as monitored by the hydrolysis of 100 µM nitrocefin.

https://doi.org/10.1371/journal.pone.0056926.g002

thumbnail
Table 2. Concentration of the inhibitors required to reduce the enzyme activity by 50%.

https://doi.org/10.1371/journal.pone.0056926.t002

Among the four inhibitors studied, tazobactam displayed lowest IC50 value (1 nM), closely followed by clavulanic acid (14 nM). Comparatively, the IC50 value of ZINC03787097 is quite high (180 nM), but it is similar to that of sulbactam (212 nM).

Steady-state Kinetic Parameters Determination

The steady-state kinetics of the purified CTX-M-15 was carried out on nitrocefin, ampicillin, piperacillin, cefazolin, cefuroxime, cefotaxime, ceftriaxone, ceftazidime, cefepime and aztreonam. The representative Michaelis-Menten plots are shown in figure S2, and the deduced kinetic parameters (kcat, Km and kcat/Km) are summarized in table 3.

thumbnail
Table 3. Steady-state kinetic parameters of CTX-M-15 from Enterobacter cloacae.

https://doi.org/10.1371/journal.pone.0056926.t003

Kinetic parameters of the purified CTX-M-15 β-lactamase revealed a hydrolytic profile that is a characteristic of molecular class A CTX-M type β-lactamases and had activity against restricted as well as expanded spectrum β-lactam antibiotics [36]. The enzyme showed good affinity (Km in 15–60 µM range) for all the studied antibiotics except ceftazidime and cefepime, for which Km was 1209 and 548 µM, respectively. Low affinity along with poor catalytic activity (1.51 and 17 s−1 for ceftazidime and cefepime, respectively) made CTX-M-15 inefficient in hydrolyzing these antibiotics. The representative catalytic efficiency for ceftazidime and cefepime are 0.0012 and 0.03 µM−1 s−1, respectively. It was also clear from table 3 that the enzyme had high catalytic efficiency (kcat/Km in 2–4 µM−1 s−1 range) for ampicillin, piperacillin, cefazolin, cefuroxime, cefotaxime and ceftriaxone. The highest catalytic activity and efficiency was observed for nitrocefin, the values being 582 s−1 and 16.6 µM−1 s−1, respectively. The study clearly indicates that CTX-M-15 had some residual activity, although relatively low, on cefepime and ceftazidime, which was further confirmed from MICs of blaCTX-M-15 expressing E. coli DH5α transformed cells.

The enzyme kinetics parameters of CTX-M-15 in the presence of various inhibitors are presented in table 4. CTX-M-15 was efficiently acylated by clavulanic acid, sulbactam and tazobactam, and individual kinetic parameters were computed. The catalytic activity of CTX-M-15 on clavulanic acid, sulbactam and tazobactam (measured as k+3) was determined as 0.0019, 0.0038 and 0.0057 s−1, respectively. Moreover, the affinity (Ki) of CTX-M-15, determined at steady-state, was found to be highest for tazobactam (0.017 µM) as compared to clavulanic acid (0.099 µM) and sulbactam (0.062 µM). The calculated Ki values (Ki = k+3K/k+2) for clavulanic acid, sulbactam and tazobactam were 0.106, 0.131 and 0.013 µM, respectively, and was in good agreement with the experimentally determined values (Table 4). The Ki of CTX-M-15 for ZINC03787097 was found to 0.388 µM, which was much higher than the other inhibitors. The acylation efficiency (k+2/K) of the enzyme was determined from the slope of a plot of ki values versus different inhibitor concentrations (Figure 3). It was found to be highest for tazobactam (0.44 µM−1 s−1), followed by sulbactam (0.029 µM−1 s−1) and clavulanic acid (0.018 µM−1 s−1). The high acylation efficiency and low deacylation efficiency (measured as k+3) made tazobactam best among mechanism based inhibitors.

thumbnail
Figure 3. Effect of various inhibitors on the kinetics of CTX-M-15.

Variation of inactivation rate constant (ki) versus different inhibitor concentrations for CTX-M-15 from clinical Enterobacter cloacae strain EC-15. The dotted line represents the best fit (linear regression) among the data points.

https://doi.org/10.1371/journal.pone.0056926.g003

thumbnail
Table 4. Effect of various inhibitors on the steady-state kinetic parameters of CTX-M-15.

https://doi.org/10.1371/journal.pone.0056926.t004

Mechanism of Action of ZINC03787097

Figure S3 shows Lineweaver-Burg plot for the hydrolysis of different concentrations of nitrocefin by CTX-M-15 in the presence of various ZINC03787097 concentrations. The values of Km for nitrocefin hydrolysis in the presence of 0.00, 0.25, 0.50 and 0.75 µM ZINC03787097 were 42, 67, 94 and 96 µM, respectively. On the other hand, the value of Vmax ( = 3.22×10−7 M s−1) was similar in all cases. Thus, it is clear that ZINC03787097 inhibited CTX-M-15 reversibly through competitive inhibition.

Stability of ZINC03787097 in the Presence of CTX-M-15

The absorption spectrum of ZINC03787097 gave characteristic peaks at 283 nm and 222 nm even after pre-incubation with CTX-M-15 for 3600 seconds. Moreover, no detectable hydrolysis of ZINC03787097 by CTX-M-15, as monitored by measuring the change in absorbance at 283 nm and 222 nm, was detected even after 3600 seconds incubation (Figure S4).

In-silico Screening of non-β-lactam Based β-lactamase Inhibitor and its Evaluation

Fitness score from GOLD program of clavulanic acid, sulbactam and tazobactam was 34.89, 34.36 and 41.66 respectively whereas fitness score of screened compound (ZINC03787097) was 47.56. It is clear that the compound ZINC03787097 was found to have highest fitness score as compared to that of known inhibitors of CTX-M-15.

Three hydrogen bonds were observed in between the novel compound and CTX-M-15. Amino acids Tyr-108, Ser-133 and Asn-173 were involved in hydrogen bonding interactions (Figure 4). Comparatively, lesser hydrogen bond interactions were observed in the case of CTX-M complexed with known inhibitors. Moreover, amino acids which are important for catalytic activity (Ser-73, Ser-75, Tyr-108, Ser-133, Asn-135, Thr-218, Gly-239 and Ser-240) were found to interact with ZINC03787097 through hydrophobic interactions (Figure 4).

thumbnail
Figure 4. Molecular docking of various inhibitors at the active site of CTX-M-15.

Panel (A) shows molecular docking of clavulanic acid (red), sulbactam (green), tazobactam (yellow) and screened compound ZINC03787097 (blue) at the active site of CTX-M-15 from clinical strain of Enterobacter cloacae (EC-15). Panel (B) represents close view of interacting amino acids in which dashed lines represent hydrogen bonds. Ambler positions of amino acids are Ser-73 (Ser-42), Tyr-108 (Tyr-77), Ser-133 (Ser-102), Asn-135 (Asn-104), Asn-173 (Asn-141), Gly-239 (Gly-208), Ser-240 (Ser-209), Thr-218 (Thr-188).

https://doi.org/10.1371/journal.pone.0056926.g004

Discussion

The treatment of bacterial infections with antibiotics is one of the key concepts of human medicine. However, the effectiveness of antibiotics has become limited owing to an increase in bacterial antibiotic resistance, which represents a global health problem with a strong social and economic impact [37], [38]. The resistance to β-lactam antibiotics, including penicillins and cephalosporins, which are amongst the most widely used class of antibiotics, is a serious problem and needs an immediate attention. Several factors contribute to this resistance mechanism like (i) mutations in the target of these drugs i.e. penicillin binding proteins involved in cell-wall biosynthesis [39], (ii) deletion and/or modification of the porin channels through which the drugs diffuse [40], (iii) expression of pumps that export the drugs out of the bacterial cells [41], and (iv) overexpression of β-lactamases in the presence of antibiotics [42], [43]. The most widespread resistance mechanism remains the expression of β-lactamase enzymes, which inactivate the β-lactam antibiotics by hydrolyzing their β-lactam ring [44], [45].

In India, CTX-M-15 is the most widely disseminated ESBL in hospital and community settings. It is highly active on cefotaxime and increasing resistance has been observed against ceftazidime as well, suggesting that the enzyme is evolving as a result of ceftazidime selection pressure [1], [9]. In this study, we have cloned blaCTX-M-15 from an Enterobacter cloacae strain (EC-15) of clinical origin (Genbank accession no.: JN860195.1) and the enzyme was purified to homogeneity (Figure S1). The resistance pattern was observed by determining MICs of common β-lactam antibiotics on E. coli DH5α transformed with cloned blaCTX-M-15 gene and steady-state enzyme kinetics of CTX-M-15 with β-lactam antibiotics and inhibitors. This is the first study that reported detailed CTX-M-15 inhibition kinetics mediated by conventional β-lactamase inhibitors such as clavulanic acid, sulbactam and tazobactam. An attempt has also been made to identify novel non-β-lactam core containing β-lactamase inhibitors by performing virtual screening of ZINC database. One such inhibitor, ZINC03787097, was identified and characterized by MICs, IC50 and enzyme kinetics.

MICs analysis showed that Enterobacter cloacae clinical isolate EC-15 is highly resistant to narrow as well as extended spectrum β-lactam antibiotics such as ampicillin, penicillin and various cephalosporins including cefotaxime and ceftazidime (Table 1). The MIC values obtained in our study against different β-lactam antibiotics were in excellent agreement with an earlier report [46]. Although, the MIC values for ceftazidime, cefepime and aztreonam were comparatively lower, E. coli DH5α cells expressing cloned blaCTX-M-15 were also resistant to these antibiotics. The lower MICs for ceftazidime, cefepime and aztreonam observed in our study is due to less selection pressure on blaCTX-M-15 harbouring bacterial population, as the consumption of these antibiotics in the treatment of bacterial infections is low. The MIC results were well supported by steady-state kinetics data, which revealed that CTX-M-15 possessed high activity against various cephalosporins including cefotaxime, and had ceftazidime, cefepime and aztreonam cleavage activity as well (Figure S2, Table 2). The broad spectrum β-lactamase activity of CTX-M-15 on various antibiotics can be explained by the replacement of Asp-240 by Gly-240, as it was evolved from CTX-M-3. Glycine at position 240 improved the substrate spectrum to include ceftazidime and aztreonam as well. The mutation of the active-site (but non-catalytic) Asp-240 to Gly-240 abolished the interaction between Asp-240 and Asn-270, which in turn improved the flexibility of the C-terminal β3 strand of the enzyme, and thus expanded the substrate spectrum to include ceftazidime and aztreonam [47]. The improvement in ceftazidime hydrolysis due to Asp240Gly substitution in CTX-M enzymes had also been reported in CTX-M-27 and CTX-M-16, which were evolved from CTX-M-9 and CTX-M-14, respectively [48], [49]. Overall, the kinetic data together with MIC values clearly showed that the β-lactamase included in this study is a typical CTX-M-15 enzyme, effective in hydrolyzing β-lactam antibiotics. It possessed high cefotaxime hydrolyzing capability and is also able to hydrolyze ceftazidime as well.

The combination of antibiotics with β-lactam based inhibitors (such as clavulanic acid, sulbactam and tazobactam) is very effective in controlling the widespread dissemination of β-lactamases. In this study, the effect of β-lactamase inhibitors on CTX-M-15 was studied by measuring MICs of various β-lactam antibiotics in combination with different inhibitors and by determining enzyme inhibition kinetic parameters. It was found that tazobactam reduced the MICs of β-lactam antibiotics to 2–16 folds and was the most potent mechanism based inhibitor of CTX-M-15 followed by clavulanic acid and sulbactam. Moreover, we found that the MICs of antibiotics of cephalosporin group were also reduced in the presence of ZINC03787097 to a level observed in the case of antibiotic-sulbactam combination (Table 1). The major cause of concern to the currently available β-lactam-β-lactamase inhibitor formulations is the emergence of resistance to amoxicillin-clavulanate and ticarcillin-clavulanate in isolates of E. coli and K. pneumoniae [50][53]. The emergence of this phenotype might be resulted from the production of β-lactamases that are not susceptible to inhibitors (e.g., Amp C from Enterobacter spp, or P. aeruginosa or metallo-β-lactamases) or the over-expression of β-lactamases due to mutations in the promoter region of the gene and/or high copy number of plasmids carrying the bla gene, as in the case of TEM-1 [54][56]. Thus, ZINC03787097 in combination with antibiotics of cephalosporin group could be used as an alternative to the conventional antibiotic-inhibitor combinations.

The enzyme inhibition kinetics (Figure 3, Table 4) revealed that CTX-M-15, like other class A serine β-lactamases, hydrolyses β-lactam antibiotics by a two step mechanism: acylation and deacylation. The acylation step is mediated by the nucleophilic attack of an active-site serine O-atom on the carbonyl carbon of the β-lactam ring, resulting in formation of covalent acyl-enzyme. On the other hand, deacylation is achieved by the activation of a water molecule by Glu-166, which acts as a general base. The resulting hydroxide then attacks the acyl carbonyl and hydrolyzes the acyl-enzyme intermediate [57]. In the case of interaction with β-lactam antibiotics, the deacylation rate is very high and the active enzyme is rapidly regenerated, whereas, interaction with β-lactam inhibitors results in an acyl-enzyme complex of substantial stability. We found that tazobactam had greatest affinity among conventional inhibitors for CTX-M-15 and displayed highest acylation efficiency and lowest deacylation efficiency. Moreover, the low Ki value (0.388 µM) and negligible hydrolysis (kcat) of inhibitor ZINC03787097 was observed (Figure S4). It competes with β-lactam antibiotic substrates for the active site of CTX-M-15 (competitive inhibition) and binds tightly through hydrogen bonding and hydrophobic interactions with the key residues involved in catalysis (Figure 4).

Conclusion

Our study concludes that ZINC03787097 is a novel non-β-lactam inhibitor that complements the active site of the enzyme and interacts with key residues involved in β-lactam recognition and hydrolysis. Unlike conventional β-lactamase inhibitors, it binds to the active site through non-covalent interactions (hydrogen bonding and hydrophobic interactions). The advantage of using such inhibitors is that β-lactamase producers cannot recruit the existing resistance mechanisms against them. Our kinetic studies together with IC50 values and promising MICs results (when used in combination with antibiotics of cephalosporin group) indicate that ZINC03787097 is a suitable lead molecule for the development of more potent non-β-lactam based β-lactamase inhibitors.

Supporting Information

Figure S1.

Cloning and purification of CTX-M-15. Panel (A) shows the vector used for cloning and expression of blaCTX-M-15 gene. Panel (B) is the SDS-PAGE of the purified CTX-M-15. Lane 1 and 2 are overexpressed total cell protein and purified protein, respectively. The single band represents molecular mass of 31 kDa.

https://doi.org/10.1371/journal.pone.0056926.s001

(TIF)

Figure S2.

Steady-state kinetics of CTX-M-15 on various β-lactam antibiotics. The hydrolysis of good substrates was analysis by plotting graphs according to Michaelis-Menten equation and the kinetic parameters (kcat and Km) were determined. For poor substrates, Km was determined as Ki in a competition experiment using 100 µM nitrocefin as reporter substrate, and kcat was estimated from the initial rates observed under saturating substrate concentrations (zeroth order reaction kinetics). Each point on these graphs is the mean value obtained from three independent experiments.

https://doi.org/10.1371/journal.pone.0056926.s002

(TIF)

Figure S3.

Lineweaver-Burg plot. The mechanism by which ZINC03780797 inhibits CTX-M-15 as determined by monitoring the hydrolysis of different concentrations of nitrocefin (10–100 µM) by 0.50 nM CTX-M-15 in the presence of varying ZINC03787097 concentrations: (a) 0 µM, (b) 0.25 µM, (c) 0.50 µM, and (d) 0.75 µM.

https://doi.org/10.1371/journal.pone.0056926.s003

(TIF)

Figure S4.

Stability of ZINC03787097 in the presence of CTX-M-15. Panel (A) shows the absorbance spectra of ZINC03787097 before and after 3600 seconds pre-incubation with CTX-M-15 at 30°C. Panel (B) shows the hydrolysis of ZINC03787097 by CTX-M-15 at 30°C monitored by measuring the change in absorbance at 283 nm and 222 nm.

https://doi.org/10.1371/journal.pone.0056926.s004

(TIF)

Author Contributions

Conceived and designed the experiments: AUK MTR. Performed the experiments: MF MTR. Analyzed the data: AUK MTR MF MD. Contributed reagents/materials/analysis tools: AUK. Wrote the paper: MTR AUK.

References

  1. 1. Bonnet R (2004) Growing Group of Extended-Spectrum β-Lactamases: the CTX-M Enzymes. Antimicrob. Agents Chemother. 48: 1–14.
  2. 2. Ambler RP, Coulson AF, Frere JM, Ghuysen JM, Joris B, et al. (1991) A standard numbering scheme for the class A β-lactamases. Biochem. J. 276: 269–270.
  3. 3. Bush K, Jacoby GA, Medeiros AA (1995) A functional classification scheme for beta-lactamases and its correlation with molecular structure. Antimicrob. Agents Chemother. 39: 1211–1233.
  4. 4. Coque TM, Baquero F, Canton R (2008) Increasing prevalence of ESBL-producing Enterobacteriaceae in Europe. Euro Surveill. 13(47): pii_19044. http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId_19044.
  5. 5. Hawkey PM, Jones AM (2009) The changing epidemiology of resistance. J. Antimicrob. Chemother. 6: i3–i10.
  6. 6. Naas T, Oxacelay C, Nordmann P (2007) Identification of CTX-M type extended-spectrum-beta-lactamase genes using real-time PCR and pyrosequencing. Antimicrob. Agents Chemother. 51: 223–230.
  7. 7. Pérez-Llarena FJ, Kerff F, Abián O, Mallo S, Fernández MC, et al. (2011) Distant and new mutations in CTX-M-1 β-lactamase affect cefotaxime hydrolysis. Antimicrob. Agents Chemother. 55: 4361–4368.
  8. 8. Bauernfeind A, Casellas JM, Goldberg M, Holley M, Jungwirth R, et al. (1992) A new plasmidic cefotaximase from patients infected with Salmonella typhimurium. Infection 20: 158–163.
  9. 9. Karim A, Poirel L, Nagarajan S, Nordmann P (2001) Plasmid mediated extended-spectrum β-lactamase (CTX-M-3 like) from India and gene association with insertion sequence ISEcp1. FEMS Microbiol. Lett. 201: 237–241.
  10. 10. Ishii Y, Ohno A, Taguchi H, Imajo S, Ishiguro M, et al. (1995) Cloning and sequence of the gene encoding a cefotaxime-hydrolyzing class A β-lactamase isolated from Escherichia coli. Antimicrob. Agents Chemother. 39: 2269–2275.
  11. 11. Gniadkowski M, Schneider I, Palucha A, Jungwirth R, Mikiewicz B, et al. (1998) Cefotaxime-resistant Enterobacteriaceae isolates from a hospital in Warsaw, Poland: identification of a new CTX-M-3 cefotaxime-hydrolyzing β-lactamase that is closely related to the CTX-M-1/MEN-1 enzyme. Antimicrob. Agents Chemother. 42: 827–832.
  12. 12. Paterson DL, Bonomo RA (2005) Extended-spectrum β- lactamases: a clinical update. Clin. Microbiol.Rev. 18: 657–686.
  13. 13. Shakil S, Khan AU (2010) Infected foot ulcers in male and female diabetic patients: a clinico-bioinformative study. Ann. Clin. Microbiol. Antimicrob. 9: 2–11.
  14. 14. Bush K (2010) Alarming β-lactamase-mediated resistance in multi-drug resistant Enterobacteriaceae. Curr. Opin. Microbiol. 13: 558–564.
  15. 15. Bush K (2010) Bench-to-bedside review: the role of β-lactamases in antibiotic-resistant Gram-negative infections. Crit. Care 14: 224.
  16. 16. Bethal CR, Taracila M, Shyr T, Thomson JM, Distler AM, et al. (2011) Exploring the inhibition of CTX-M-9 by β-lactamase inhibitors and carbapenems. Antimicrob. Agents Chemother. 55: 3465–3475.
  17. 17. Meroueh SO, Roblin P, Golemi D, Maveyraud L, Vakulenko SB, et al. (2002) Molecular dynamics at the root of expansion of function in the M69L inhibitor-resistant TEM β-lactamase from Escherichia coli. J. Am. Chem. Soc. 124: 9422–9430.
  18. 18. Sun T, Bethel CR, Bonomo RA, Knox JR (2004) Inhibitor-resistant class A β-lactamase: Consequences of the Ser130-to-Gly mutation seen in apo and Tazobactam structures of the SHV-1 variant. Biochemistry 43: 14111–14117.
  19. 19. Bush K (1999) Beta-lactamases of increasing clinical importance. Curr. Pharm. Des. 5(11): 839–845.
  20. 20. Hanson ND, Sanders CC (1999) Regulation of inducible AmpC beta-lactamase expression among Enterobacteriaceae. Curr. Pharm. Des. 5(11): 881–894.
  21. 21. Powers RA, Blazquez J, Weston GS, Morosini MI, Baquero F, et al. (1999) The complex structure and antimicrobial activity of a non-beta-lactam inhibitor of AmpC beta-lactamase. Protein Sci. 8(11): 2330–2337.
  22. 22. Power RA, Morandi F, Shoichet BK (2002) Structure-based discovery of a novel, noncovalent inhibitor of AmpC β-lactamase. Structure 10: 1012–1023.
  23. 23. Shakil S, Akram M, Ali SM, Khan AU (2010) Acquisition of extended-spectrum β-lactamase producing Escherichia coli strains in male and female infants admitted to a neonatal intensive care unit: molecular epidemiology and analysis of risk factors. Journal of Molecular Medicine 59: 948–954.
  24. 24. Hasan S, Danishuddin M, Adil M, Singh K, Verma PK, et al. (2012) Efficacy of E. officinalis on the Cariogenic Properties of Streptococcus mutans: A Novel and Alternative Approach to Suppress Quorum-Sensing Mechanism. PLoS ONE 7(7): e40319
  25. 25. Clinical and Laboratory Standards Institute (2011) Performance standards for antimicrobial susceptibility testing: 21st informational supplement. M100-S21. CLSI, Wayne, PA.
  26. 26. Djamdjin L, Naas T, Tande D, Cuzon G, Hanrotel-Saliou C, et al. (2011) CTX-M-93, a variant lacking penicillin hydrolytic activity. Antimicrob. Agents Chemother. 55(5): 1861–1866.
  27. 27. Stachyra T, Pechereau M-C, Bruneau J-M, Claudon M, Frere J-M, et al. (2010) Mechanistic studies of the inactivation of TEM-1 and P99 by NXL104, a novel non-β-lactam β-lactamase inhibitor. Antimicrob. Agents Chemother. 54(12): 5132–5138.
  28. 28. Galleni M, Franceschini N, Quinting B, Fattorini L, Orefici G, et al. (1994) Use of chromosomal class A β-lactamase of Mycobacterium fortuitum D316 to study potentially poor substrates and inhibitory β-lactam compounds. Antimicrob. Agents Chemother. 38: 1608–1614.
  29. 29. Rehman MT, Dey P, Hassan MI, Ahmad F, Batra JK (2011) Functional role of Glutamine 28 and Arginine 39 in double stranded RNA cleavage by Human Pancreatic Ribonuclease. PLoS ONE 6(3): e17159
  30. 30. Vilar M, Galleni M, Solmajer T, Turk B, Frere JM, et al. (2001) Kinetic study of two novel enantiomeric tricyclic β-lactams which efficiently inactivate class C β-lactamases. Antimicrob. Agents Chemother. 45: 2215–2223.
  31. 31. Matagne A, Ghuysen MF, Frere JM (1993) Interaction between active-site serine β-lactamases and mechanism-based inactivators: a kinetic study and an overview. Biochem. J. 295: 705–711.
  32. 32. Waley SG (1992) β-lactamase: mechanism of action, p.198–228. In M I Page (ed.), The chemistry of β-lactams. Blackie A and P, London, United Kingdom.
  33. 33. Brooks BR, Bruccoleri RE, Olafson BD, Sate DJ, Swaminathan S, et al. (1983) CHARMM: A program for macromolecular energy, minimization, and dynamics calculations. J. Comp. Chem. 4: 187–217.
  34. 34. Irwin JJ, Shoichet BK (2005) ZINC–a free database of commercially available compounds for virtual screening. J. Chem. Inf. Model. 45: 177–182.
  35. 35. Jones G, Willett P, Glen RC (1995) Molecular recognition of receptor sites using a genetic algorithim with a description of desolvation. J. Mol. Biol. 245: 43–53.
  36. 36. Bonnet R, Sampanio JL, Labia R, De-Champs C, Sirot D, et al. (2000) A novel CTX-M β-lactamase (CTX-M-8) in cefotaxime-resistant Enterobacteriacae isolated in Brazil. Antimicrob. Agents Chemother. 44: 1936–1942.
  37. 37. Rice LB, Bonomo RA (2000) beta-Lactamases: Which ones are clinically important? Drug Resist. Updates. 3(3): 178–189.
  38. 38. Cosgrove S, Carmeli Y (2003) The impact of antimicrobial resistance on health and economic outcomes. Clin. Infect. Dis. 36: 1433–1437.
  39. 39. Chambers HF (1999) Penicillin-binding protein-mediated resistance in Pneumococci and Staphylococci. J. Infect. Dis. (Suppl. 2): 353–359.
  40. 40. Li XZ, Nikaido H (2004) Efflux-mediated drug resistance in bacteria. Drugs 64(2): 159–204.
  41. 41. Li XZ, Zhang L, Nikaido H (2004) Efflux pump-mediated intrinsic drug resistance in Mycobacterium smegmatis. Antimicrob. Agents Chemother. 48(7): 2415–2423.
  42. 42. Thiolas A, Bornet C, Davin-Regli A, Pages JM, Bollet C (2004) Resistance to Imipenem, Cefepime, and Cefpirome associated with mutation in Omp36 osmoporin of Enterobacter aerogenes. Biochem. Biophys. Res. Commun. 317(3): 851–856.
  43. 43. Tondi D, Morandi F, Bonnet R, Costi MP, Shoichet BK (2005) Structure-based optimization of a non-β-lactam lead results in inhibitors that do not up-regulate β-lactamase expression in cell culture. J. Am. Chem. Soc. 127: 4632–4639.
  44. 44. Bennett PM, Chopra I (1993) Molecular basis of beta-lactamase induction in bacteria. Antimicrob. Agents Chemother. 37(2): 153–158.
  45. 45. Essack SY (2001) The development of beta-lactam antibiotics in response to the evolution of beta-lactamases. Pharm. Res. 18(10): 1391–1399.
  46. 46. Poirel L, Gniadkowski M, Nordmann P (2002) Biochemical analysis of the ceftazidime-hydrolysing extended-spectrum β-lactamase CTX-M-15 and of its structurally related β-lactamase CTX-M-3. J. Antimicrob, Chemother. 50: 1031–1034.
  47. 47. Celenza G, Luzi C, Aschi M, Segatore B, Setacci D, et al. (2008) Natural D240G Toho-1 mutant conferring resistance to ceftazidime: biochemical characterization of CTX-M-43. J. Antimicrob. Chemother. 62: 991–997.
  48. 48. Bonnet R, Dutour C, Sampaio JLM, Chanal C, Sirot D, et al. (2001) Novel cefotaximase (CTX-M-16) with increased catalytic efficiency due to substitution Asp240Gly. Antimicrob. Agents Chemother. 45: 2269–2275.
  49. 49. Bonnet R, Recule C, Baraduc R, Chanal C, Sirot D, et al. (2003) Effect of D240G substitution in a novel ESBL CTX-M-27. J. Antimicrob. Chemother. 52: 29–35.
  50. 50. Drawz SM, Bonomo RA (2010) Three decades of beta-lactamase inhibitors. Clin. Microbiol. Rev. 23: 160–201.
  51. 51. Martinez JL, Cercenado E, Rodriguez-Creixems M, Vicente-Perez MF, Delgado-Iribarren A, et al. (1987) Resistance to β-lactam/clavulanate. Lancet ii: 1473.
  52. 52. Martinez JL, Vicente-Perez MF, Delgado-Iribarren A, Perez-Diaz JC, Baquero F (1989) Small plasmids are involved in amoxicillin-clavulanate resistance in Escherichia coli. Antimicrob. Agents Chemother. 33: 595.
  53. 53. Sanders CC, Iaconis JP, Bodey GP, Samonis G (1988) Resistance to ticarcillin-potassium clavulanate among clinical isolates of the family Enterobacteriaceae: role of PSE-1 β-lactamase and high levels of TEM-1 and SHV-1 and problems with false susceptibility in disk diffusion tests. Antimicrob. Agents Chemother. 32: 1365–1369.
  54. 54. Williams H, King A, Shannon K, Phillips I (1988) Amoxicillin/clavulanate resistant Escherichia coli. Lancet i: 304–305.
  55. 55. Wu PJ, Shannon K, Phillips I (1994) Effect of hyperproduction of TEM-1 β-lactamase on in vitro susceptibility of Escherichia coli to β-lactam antibiotics. Antimicrob. Agents Chemother. 38: 494–498.
  56. 56. Wu PJ, Shannon K, Phillips I (1995) Mechanisms of hyperproduction of TEM-1 β-lactamase by clinical isolates of Escherichia coli. J. Antimicrob. Chemother. 36: 927–939.
  57. 57. Strynadka NCJ, Adachi H, Jensen SE, John K, Sielecki A, et al. (1992) Molecular structure of the acyl-enzyme intermediate in β-lactam hydrolysis at 1.7 Å resolution. Nature 359: 700–705.